syllable, expressing the general and indeterminate idea of 'calling,' and found to occur in connected speech only when limited and defined by the suffixes which are attached to it. This is not, however, a peculiarity which can exempt the words so formed from a like treatment, leading to like conclusions, with the rest; we must still trust in the reality of our analysis; and especially, when we consider such forms as the Sanskrit vak-mi, vak-shi, vak-ti, where the mi, shi, and ti are recognizable pronouns, making compounds which mean clearly 'call-I,' 'call-thou,' 'call-he,' we cannot doubt that the element voc (vak) had also once an independent status, that it was a word, a part of spoken speech, and that the various forms which contain it were really produced by the addition of other elements to it, and their fusion together into a single word, in the same manner in which we have fused truth and full into truthful, truth and loose into truthless, true and like into truly.
The same conclusion may be stated in more general terms, as follows. The whole body of suffixes, of formative endings, is divided into two principal classes: first, primary, or such as form derivatives directly from roots; second, secondary, or such as form derivatives from other derivatives, from themes containing already a formative element. But the difference between these two classes is in their use and application, not in their character and origin. No insignificant portion of each is traceable back to independent words, and the presumption alike for each is that in all its parts it was produced in the same manner. If, then, we believe that the themes to which the secondary endings are appended were historical entities, words employed in actual speech before their further composition, we must believe the same respecting the roots to which are added the primary endings: these are not less historical than the others.
The conclusion is one of no small consequence. Elements like voc, each composing a single syllable, and containing no traceable sign of a formative element, resisting all our attempts at reduction to a simpler form, are what we arrive at as the final results of our analysis of the Indo-European vocabulary; every word of which this is made up—save those
